In this chapter, I present the results obtained through latent variable structural equations modeling using SmartPLS (Ringle, Wende, and Will, 2005) and covariancebased AMOS 7.0 (Arbuckle, 1983–2006). The PLS model is estimated with reference to three different data sets: (1) the pooled data set of 241 cases consisting of 104 aggregated multiple responses and 137 single responses (referred to as D1), (2) the aggregated multiple responses dataset comprising 104 cases (D2), and (3) the primary responses dataset of 241 single responses (D3). The comparison of the estimates for the three datasets enables an evaluation of the impact of using multiple respondents. The AMOS model was estimated only for the pooled data set (D1) and only for research question one. After briefly describing the data preparation procedure in section 5.1, I evaluate the measurement models for each of the three data sets in section 5.2. Next, in section 5.3, I present the parameter estimates for the structural model that corresponds to research question one. I also examine the differences in estimation results obtained from the three data sets in this section. Ultimately, I present the parameter estimates for research question two in section 5.4.


Relationship Quality Path Coefficient Direct Communication Brand Image Formative Indicator 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2012

Authors and Affiliations

  • Stefan Worm

There are no affiliations available

Personalised recommendations